PV Pak Preview

Status
Not open for further replies.

Laredo7mm

Senior Member
ECF Veteran
Mar 20, 2009
154
1
I always use sockets/holders for ICs - solder the socket and connect the IC into it after it's cooled down.

Yeah, I looked for a socket for the IC that I am using bit I couldn't find one. The chip is a soic16 and has 16, 0.5mm wide pins that 1.27mm on center. I looked at mouser and digikey and the only thing I found were huge and expensive.

I have 3 IC's left of the original 5 FREE samples. Man I love free samples! :thumb: I just ordered some more wire to board connectors and some different capacitors. The original two pin connectors I ordered were really small. Not bad soldering them to the board, but the associated crimp on pins are a nightmare to work with.

Unfortunately, I am not sure if the touch switch is going to work. I think I misunderstood the way the IC works. Basically the IC is active or not. When active, Vout is 5V which is what I want. But when the IC is off, the output is battery. So basically, I can set the output voltage at 3.7V or 5V with the touch switch.

I am going take a look at mosfets and see if I can get that to work.
 
Yeah, I looked for a socket for the IC that I am using bit I couldn't find one. The chip is a soic16 and has 16, 0.5mm wide pins that 1.27mm on center. I looked at mouser and digikey and the only thing I found were huge and expensive.

I have 3 IC's left of the original 5 FREE samples. Man I love free samples! :thumb: I just ordered some more wire to board connectors and some different capacitors. The original two pin connectors I ordered were really small. Not bad soldering them to the board, but the associated crimp on pins are a nightmare to work with.

Unfortunately, I am not sure if the touch switch is going to work. I think I misunderstood the way the IC works. Basically the IC is active or not. When active, Vout is 5V which is what I want. But when the IC is off, the output is battery. So basically, I can set the output voltage at 3.7V or 5V with the touch switch.

I am going take a look at mosfets and see if I can get that to work.

Something like this :

touch-switch-using-ujt-2n3819-250x169.jpg

Source: Touch Switch using UJT 2N3819 | Circuit Project Electronic

But to achieve auto-off I'd remove the 1.5V cell and just connect a 3.3M res from ground to the base instead. Would need to check if the atty coil can be used in place of the relay (if the mosfet can handle 1A). Best to check what the off current is. Maybe used with a minitiature relay would be best.
 
Last edited:
Would I need a protection diode since I am running an inductive load to protect against the current backflow when the power is cut?
That's wise, although perhaps unnecessary with a low power relay. Make sure it's connected the right way around or the tranny might blow. An atty coil would not cause a problem.
 
I have no idea what you guys are talking about (lol), but it looks very nice!! :)

It's just that a base current overload, if beyond the normal functional limit when considering the device's current gain parameter might disturb the semiconductor ion exchange property causing malfunction - that's all.
 
Last edited:
Here is the latest PCB design:

PCB_Mosfet.jpg


The M1 component ubove the main IC is the Mosfet. Good for 11A.

Not sure about the red and green ;)

Seriously, are you sure M1 is the mosfet?

edit (add): Yes, I can see from the circuit that it is. But why 8 legs shown? You will connect it in a DIL socket?
 
Last edited:
Laredo7mm - I have been thinking - and 2 things:

Putting the load (coil or relay) on the collector side r/t/ emitter side would be helpful wrt base bias; easier 'on'.

I think a 2 transistor design would be better. Emitter follower pattern. A good choice for the first transistor being a FET.

I will be putting a touch switch into the Touch / Easy Puff - though on the mouthpiece.
 

Laredo7mm

Senior Member
ECF Veteran
Mar 20, 2009
154
1
...Putting the load (coil or relay) on the collector side r/t/ emitter side would be helpful wrt base bias; easier 'on'.

I think a 2 transistor design would be better. Emitter follower pattern. A good choice for the first transistor being a FET...quote]

Like a Darlington Pair type set up?

In the circuit I have the load on the drain side so the mosfet is acting like a switch closing the circuit to ground. That way I can avoid the forward voltage drop through the NPN.
 
...Putting the load (coil or relay) on the collector side r/t/ emitter side would be helpful wrt base bias; easier 'on'.

I think a 2 transistor design would be better. Emitter follower pattern. A good choice for the first transistor being a FET...quote]

Like a Darlington Pair type set up?

In the circuit I have the load on the drain side so the mosfet is acting like a switch closing the circuit to ground. That way I can avoid the forward voltage drop through the NPN.

Essentially, yes.

It's like one super transistor. (Interested readers can refer to here: Darlington Pairs)

The first transistor would be a small FET. A capacitor from ground to base would provide some de-bouncing (say 22nF). I'd also keep a 4m7 res from ground to base of the mosfet and play with the design on a breadboard (everyone doing any electronics should get a breadboard).
 

Laredo7mm

Senior Member
ECF Veteran
Mar 20, 2009
154
1
Yeah, I have slowed it down a bit. I think I have figured out what the issue is. With the help of some electronics forums, I was advised to make my power and power ground leads wide and short. Something about parasitic capacitance shorting out my IC. I believe it, since my board would work for a little while and then stop.

Soon, very soon, I think I will have it all sorted out. Well...maybe.
 
Status
Not open for further replies.

Users who are viewing this thread